Network connection optimization method and device for cellular communication module, medium and program product

Abstract

The invention provides a network connection optimization method and device for a cellular communication module, a medium and a program product, and the method comprises the steps: collecting parameter information representing a current network connection state, carrying out the feature processing of the parameter information, and constructing a multi-dimensional feature vector; for a plurality of candidate connection strategies, calculating a score value of each candidate connection strategy based on the multi-dimensional feature vector and a preset weight value; performing normalization processing on each score value to obtain probability distribution of candidate connection strategies; according to the probability distribution, periodically executing probability sampling once from the plurality of candidate connection strategies, determining a target connection strategy of the current period, and determining whether to perform network switching; and dynamically adjusting and calculating temperature parameters of probability distribution according to a performance evaluation result of the current connection state, and optimizing a sampling behavior of a connection strategy of a next period. According to the application, the temperature parameter is introduced as a regulation factor, so that the connection stability, the strategy switching flexibility and the overall network access quality of the cellular communication module are improved.

Description

Technical Field

[0001] This application relates to the field of wireless communication technologies, and particularly to a method, device, medium, and program product for optimizing network connections of cellular communication modules. Background Art

[0002] With the rapid development of cellular communication technologies, communication modules supporting standards such as LTE (4G) and NR (5G) have been widely used in scenarios such as industrial Internet of Things, vehicle Internet of Things, and remote control. However, in complex or dynamic network environments, the connection stability of communication modules still faces many challenges, and situations of frequent disconnection or access failure occur frequently, making it difficult to meet the actual requirements for continuous and highly reliable communication.

[0003] In practical applications, the disconnection problem of cellular communication modules is usually caused by the following factors: Frequent cell handover: In areas with strong mobility or at the edge of network coverage, the module frequently performs cell handover. If the handover fails or there is a handover delay, it may lead to connection interruption.

[0004] Severe signal quality fluctuations: Key wireless metrics such as reference signal received power (RSRP) and signal-to-noise ratio (SNR) are significantly affected by factors such as occlusion, interference, and reflection in the actual environment. Once they are in a low value for a long time or fluctuate severely, it will directly affect the link stability and data transmission quality.

[0005] Improper band selection: Cellular communication systems support multiple bands, and different bands have differences in coverage ability, interference environment, and communication capacity. If the module cannot dynamically select the optimal band according to the current scenario (for example, only relying on the default configuration to use a high-capacity but weak-penetration band), it may lead to a decline in connection performance or even disconnection; Abnormal operator network or IP channel: In some areas, there may be insufficient coverage of a single operator, or the accessed IP channel may have problems such as high load, serious packet loss, and slow connection establishment, thus affecting communication stability or causing frequent disconnection.

[0006] Therefore, the existing technologies urgently need to be improved to achieve dynamic evaluation and optimization of the access strategy of communication modules and improve connection stability in complex wireless environments. Summary of the Invention

[0007] Aiming at the deficiencies of the existing technologies, this application provides a method, device, medium, and program product for optimizing network connections of cellular communication modules, at least to solve the problems of unstable connection, lagging strategy response, and lack of adaptive decision-making mechanism of cellular communication modules in the existing technologies.

[0008] To achieve the above objectives and other advantages, some embodiments of this application provide the following aspects: In a first aspect, some embodiments of the present application provide a method for optimizing network connection for a cellular communication module, including: Real-time collecting parameter information for characterizing the current network connection status, and performing feature processing on the parameter information to construct a multi-dimensional feature vector, where the parameter information includes: base station cell handover frequency, operating frequency band type, received signal power, signal-to-noise ratio, operator network service quality index, connection stability index; For each of the constructed multiple candidate connection strategies, calculating a score value for each candidate connection strategy respectively based on the multi-dimensional feature vector and a preset weight value; Normalizing each of the score values to obtain a probability distribution of the candidate connection strategies; According to the probability distribution, performing probability sampling once periodically from the multiple candidate connection strategies to determine the target connection strategy for the current period; Determining whether the target connection strategy is consistent with the currently connected strategy of the cellular communication module. If not, controlling the cellular communication module to disconnect the current connection and establish a network access path corresponding to the target connection strategy; According to the performance evaluation result of the current connection state, dynamically adjusting the temperature parameter for calculating the probability distribution in the normalization process, so as to dynamically balance exploration and exploitation in the connection strategy selection, thereby optimizing the sampling behavior of the connection strategy for the next period.

[0009] In a second aspect, some embodiments of the present application further provide an electronic device, where the electronic device includes: One or more processors; and a memory storing computer program instructions, where the computer program instructions, when executed, cause the processors to execute the method for optimizing network connection for a cellular communication module as described in any one of the above.

[0010] In a third aspect, some embodiments of the present application further provide a computer-readable storage medium, on which computer programs and / or instructions are stored, and when the computer programs and / or instructions are executed by a processor, the method for optimizing network connection for a cellular communication module as described in any one of the above is implemented.

[0011] In a fourth aspect, some embodiments of the present application further provide a computer program product, including computer programs and / or instructions, and when the computer programs / instructions are executed by a processor, the method for optimizing network connection for a cellular communication module as described in any one of the above is implemented.

[0012] Compared with the related art, in the solution provided by the embodiments of the present application, a weighted scoring model is introduced to quantitatively evaluate multiple candidate connection strategies and perform normalization processing to form a probability distribution; a temperature parameter is introduced as a regulation factor to adjust the discreteness of the sampling probabilities among the candidate connection strategies during the scoring normalization process, realizing dynamic balance control between exploration and exploitation in the process of connection strategy selection. When the connection state is good, the system strengthens the sampling tendency of high-scoring strategies to improve connection stability; when the environment fluctuates or the signal deteriorates, the range of strategy selection is expanded to improve robustness and recovery ability, thus significantly improving the connection stability, strategy switching flexibility, and overall network access quality of the cellular communication module. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] To more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other embodiments can be obtained based on these drawings.

[0014] Figure 1 is a schematic flowchart of a network connection optimization method for a cellular communication module provided by an embodiment of the present application; Figure 2 is a system interaction flowchart of the network connection optimization method for the cellular communication module provided by an embodiment of the present application; Figure 3 is an adaptive parameter update control logic flowchart under continuous disconnection triggering provided by an embodiment of the present application; Figure 4 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0015] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0016] The First Embodiment The first embodiment of the present application relates to a network connection optimization method for a cellular communication module. Referring to Figure 1 、 Figure 2 as shown, the method may include the following steps: Step S1: Real-time collect parameter information for characterizing the current network connection status, perform feature processing on the parameter information, and construct a multi-dimensional feature vector. The parameter information includes: base station cell handover frequency, working frequency band type, received signal power, signal-to-noise ratio, operator network service quality index, and connection stability index.

[0017] Regarding step S1, specifically, the cellular communication module periodically collects key parameter information related to the network connection status through its open wireless parameter interface, and performs feature engineering processing on the collected raw data. The feature engineering processing includes: uniformly mapping the raw data with physical dimensions (such as dBm, dB, ms, etc.) to the standardized numerical interval of 0 to 1 for unified scoring, and finally combining them to form a multi-dimensional feature vector for scoring decision-making; using piecewise linear or Sigmoid functions to map some parameters to enhance the model's recognition ability for boundary intervals; introducing an exponential decay function for the cell handover frequency to generate a mobility index; introducing a moving average process for the historical disconnection times to form a stability coefficient, etc.

[0018] Furthermore, in step S1, the steps of performing feature processing on the parameter information and constructing a multi-dimensional feature vector include: Perform feature engineering processing on the original collected values of the parameter information, convert them into standardized numerical values that can be used for subsequent scoring calculations, and combine the standardized numerical values into the form of a multi-dimensional feature vector. The feature dimension of the multi-dimensional feature vector can be extended or trimmed according to the business scenario. Among them, the value range of the standardized numerical value is 0 to 1, indicating the performance of the candidate connection strategy in the parameter information dimension.

[0019] The parameter information includes the following six core features: Base station cell handover frequency: It refers to the ratio of the number of cell reselection or handover events that occur within a set time window by the cellular communication module to the time length, and is used to reflect the mobility or connection stability characteristics of the device in a specific network environment.

[0020] First, the system calculates the cell handover frequency per unit time (i.e., the handover rate), and normalizes it so that the result is limited within the interval [0,1] to prevent extreme values from interfering with the scoring. The normalized handover rate can be expressed as:

[0021] where switch_rate is the normalized handover rate; switch_count is the number of base station cell handovers that occur within the set statistical period (i.e., window_size) by the cellular communication module; window_size is the statistical time window for handover behavior.

[0022] On this basis, to further enhance the model's sensitivity to high-frequency handover (high mobility) states, the system introduces an exponential decay function to construct a mobility index. This index gradually approaches 1 as the handover rate increases, reflecting the instability of the connection state. The construction function of the mobility index is as follows:

[0023] Among them, λ is an adjustable decay coefficient that controls the response sensitivity of the function to high-frequency handovers, and the preferred value is 0.1.

[0024] Working frequency band type: It refers to the wireless communication frequency bands that the cellular communication module is currently connected to or can be connected to, such as Band3, Band8, etc., which affect signal penetration and interference characteristics.

[0025] Specifically, the basic priority table can refer to the frequency band performance characteristics configured by 3GPP. The system pre-sets a basic priority table for frequency bands, which is set according to the characteristics such as capacity, interference, and coverage ability of each frequency band in the 3GPP standard specifications. For example: Band3 (1800 MHz): The basic priority is 0.8; Band8 (900 MHz): The basic priority is 0.6; Band7 (2600 MHz): The basic priority is 0.9. For each frequency band used by a candidate connection strategy, find its basic priority value base_priority from the priority table.

[0026] To enhance the dynamic nature of priority expression, the system further corrects the priority based on the historical connection success rate suc_rate (value range 0 - 1) of the module in this frequency band. The adjusted frequency band adjustment factor adjusted_priority is calculated as follows:

[0027] Received Signal Strength Indicator (RSRP): It refers to the measured value of the receiving power of the base station downlink signal by the cellular communication module in the current connection state, with the unit of dBm, and is used to evaluate the signal strength of the wireless link.

[0028] To effectively incorporate this parameter into the connection strategy scoring model, the system performs piecewise normalization on the original RSRP data to enhance the recognition ability for different signal intervals and suppress the interference of extremely weak signals on scoring. The processing method is as follows: When the RSRP is greater than or equal to -85 dBm, it indicates that the signal quality of the environment where the module is currently located is good, and the normalization processing result is directly set to 1.0, representing the highest score for this dimension; When the RSRP is in the range of –100 dBm to –85 dBm, a linear function is used to map the signal strength to the middle section of the [0,1] range to maintain the discrimination of the score within the medium strength range. The normalization calculation method is as follows:

[0029] When the RSRP is less than –100 dBm, it indicates that the module is in a weak signal area. To avoid the scoring model being overly sensitive to extremely low values, the system introduces a scaling factor for compression mapping to weaken the scoring impact. The processing method is as follows:

[0030] The normalized eigenvalue rsrp_norm of the received signal power after processing is limited within the [0,1] range and is used as one-dimensional input in the normalized eigenvector to reflect the signal strength performance of the candidate connection strategy in the current environment.

[0031] Signal-to-noise ratio (SNR): It refers to the ratio of the useful signal power measured at the receiving end of the cellular communication module to the background noise power, with the unit of dB. It is an important parameter for evaluating the quality of the wireless channel and the anti-interference ability.

[0032] To incorporate this indicator into the scoring model and improve its discrimination, the system linearly processes the SNR and maps it to a normalized channel capacity value to quantify the performance of different connection strategies in terms of channel utilization efficiency. The specific processing method is as follows: Convert the original SNR in dB value to a linear value, and the calculation formula is:

[0033] Referring to the simplified form of the Shannon channel capacity formula, calculate the theoretical communication capacity per unit bandwidth and set the maximum capacity as the normalization upper limit (preferably 6) to obtain the normalized channel capacity:

[0034] The obtained channel capacity value capacity is used as one-dimensional input in the normalized eigenvector to participate in the scoring calculation of the candidate connection strategy, and is used to distinguish the advantages and disadvantages of different strategies in terms of anti-interference and channel resource availability dimensions.

[0035] Operator network service quality indicator: It refers to the service performance of the current cellular communication module accessing the operator (such as China Mobile, China Unicom, or China Telecom, etc.) in this network environment. This indicator comprehensively considers two dimensions of communication delay and operator's global quality evaluation, and forms a two-dimensional input feature of the scoring model after feature engineering processing.

[0036] The module periodically obtains the average round-trip delay delay_ms with the core network through network layer commands (such as ping tests). To avoid the linear error amplification of the absolute value of the time delay on the scoring model, the system uses the Sigmoid function to perform normalization mapping on it. The normalization formula is as follows:

[0037] This function uses 100ms as the median reference point to ensure that medium and low delays get higher scores, and high delays gradually compress the scores to avoid extreme deviations.

[0038] The system obtains its autonomous system number ASN according to the operator identification of the SIM card or the IP address attribution information, and queries the preset operator network performance database to return the service quality evaluation score asn_quality of the current operator. Its value range is [0,1], and the closer it is to 1, the more stable and reliable the operator is in the current area or historical statistics.

[0039] The time delay quality score delay_score and the network quality score asn_quality together constitute the operator service quality characteristics for policy scoring, and the output is a two-dimensional standardized feature, which is used to quantify the advantages and disadvantages of candidate connection policies in terms of service layer performance.

[0040] Connection stability index: It refers to that the cellular communication module constructs the master station connection stability characteristics for reflecting connection continuity and reliability by collecting its own historical behavior data during the connection process with the master station. This feature is modeled from two aspects: one is the stability coefficient calculated based on the disconnection event, and the other is the jitter penalty term based on the time delay fluctuation.

[0041] Calculation of stability coefficient (EWMA smoothing): The system counts the number of disconnections offline_count of the cellular communication module in each cycle, and uses the exponentially weighted moving average (EWMA) algorithm in combination with the previous cycle's stability score prev_stability to construct a smoothed stability evaluation index with historical memory ability. The calculation formula is as follows:

[0042] Among them, the discretization coefficient 0.1 represents the update weight of the latest observed data, and the maximum number of offline times is deducted by 5 points as the full score to ensure that the score value is in the range of [0,1]. The closer the value is to 1, the more stable the connection is.

[0043] Calculation of time delay jitter penalty term: The system calculates the standard deviation rtt_deviation of the RTT (round-trip delay) of the module's communication with the master station during the current period, which reflects the volatility of the link delay during the connection process. To limit the scoring interference caused by extreme fluctuations, a normalized jitter penalty term is introduced, and the calculation formula is:

[0044] The higher the value of this penalty term rtt_penalty, the worse the link stability, and finally it participates in the policy scoring judgment as a negative factor.

[0045] The finally output connection stability coefficient stability and the penalty term value rtt_penalty together form a two-dimensional master station connection stability feature, which are respectively used to evaluate the continuity and volatility of the connection, and assist in screening out candidate policies with better historical connection performance in the scoring model.

[0046] After completing the feature engineering processing of all parameter information, the system splices all the processed standardized feature values in a preset order to construct a multi-dimensional feature vector for candidate connection policy scoring and decision-making. In this embodiment, the feature vector contains 8 dimensions, and each dimension corresponds to the following feature parameters in turn: mobility index (mobility_index), frequency band adjustment factor (adjusted_priority), normalized feature value of received signal power (rsrp_norm), channel capacity value (capacity), delay score (delay_score), network quality score (asn_quality), connection stability coefficient (stability), and penalty term value (rtt_penalty). The above-mentioned features of each dimension are spliced into an input vector of a unified structure in a fixed order, in the form of: feature_vector = [mobility_index, adjusted_priority, rsrp_norm, capacity, delay_score, asn_quality, stability, rtt_penalty].

[0047] It should be noted that the dimensions and content of the above feature vector can be expanded or trimmed according to specific business scenarios to achieve stronger adaptability and model expansion ability.

[0048] Step S2: For the multiple candidate connection policies that have been constructed, calculate the scoring value of each candidate connection policy based on the multi-dimensional feature vector and the preset weight value.

[0049] For multiple constructed candidate connection strategies, combined with the multi-dimensional feature vectors collected and processed in the current network environment, a weighted scoring mechanism is adopted to quantitatively evaluate each candidate connection strategy. The scoring value is used to represent the comprehensive adaptability and preference degree of the strategy in the current network state.

[0050] Further, in step S2, constructing multiple candidate connection strategies specifically includes: Obtain the network access parameters that can be detected by the cellular communication module in the current network environment. The network access parameters include: cell identifier, frequency band type, operator identifier, and radio access technology type; Based on different combination methods of the network access parameters, construct multiple candidate connection strategies. The candidate connection strategies are used to represent the connection paths that the cellular communication module can select in the current network environment.

[0051] Specifically, the cellular communication module periodically scans and collects the accessible network parameter information in the current network environment through the underlying communication protocol stack and debugging interface. Such parameter information at least includes: Cell ID (Cell ID): the unique identifier of the surrounding base station cell detected by the module; Band (Band): the working frequency band to which each cell belongs, such as Band3 (1800 MHz), Band8 (900 MHz), etc.; Operator ID (Operator ID): the network operator to which the current cell belongs, such as China Mobile, China Unicom, etc., and multi-operator access can be further realized by associating with the SIM card or eSIM configuration; RAT Type (RAT Type): the supported access mode, such as LTE, NR (5G), NB-IoT, GSM, etc. The above parameters can be obtained through system information block decoding, signaling reports (such as measurement reports), frequency point scanning, etc. of the wireless module.

[0052] Base station selection dimension: such as connecting to CellA, CellB, or CellC; frequency band combination dimension: such as single-frequency access of Band3, cross-frequency carrier aggregation of Band3 + Band7; operator selection dimension: such as operator A corresponding to SIM1 and operator B corresponding to SIM2; RAT type dimension: such as independent access of 4G and 5G, or non-standalone networking (NSA) handover strategy.

[0053] Based on the various network access parameters collected above, the system generates several sets of candidate connection strategies for the module to attempt to connect according to the feasible combination strategy rules. Each set of candidate connection strategies can be regarded as a specific network access path plan. For example, in a certain network environment, the module may detect the following optional access configurations: Strategy 1: Connect to CellA (Band3, Operator A); Strategy 2: Connect to CellB (Band8, Operator B); Strategy 3: Connect to CellA + CellC (Band3+Band7 cross-frequency aggregation, Operator A).

[0054] The system takes the above three strategy combinations as the candidate connection strategies in the current cycle and enters the scoring and sampling stage to evaluate the advantages and disadvantages of each strategy in the current network state. This candidate strategy set will be dynamically updated according to the network environment, base station broadcasts, SIM configuration, and module movement status to ensure a wide coverage and strong adaptability of the strategy space, providing a rich decision-making basis for subsequent feature vector-based scoring.

[0055] Step S3: Normalize each score value to obtain the probability distribution of the candidate connection strategies.

[0056] Specifically for Step S3, assume that there are n candidate connection strategies in the current cycle, corresponding to score values S1, S2,..., S n , and the system uses a normalization function to transform these score values to obtain the sampling probability P i of each strategy, reflecting the likelihood of being selected in the current environment.

[0057] Preferably, the system uses a Softmax function with adjustable dispersion for normalization processing, and its function form is as follows:

[0058] where P i is the sampling probability of the i-th candidate connection strategy; S i is the score value of the i-th candidate connection strategy; β is the temperature parameter, which is used to control the smoothness or deviation of the normalized distribution.

[0059] The temperature parameter β is a key adjustment coefficient in the Softmax function, and its value determines the probability distribution form after score normalization. The mechanism of action of β is as follows: When the value of β approaches 0, the difference between scores is greatly amplified in the exponential function, and the normalization result tends to be average (entropy increase), that is, each strategy is given a relatively similar sampling probability, and the system shows a uniform exploration behavior, which is suitable for use in scenarios where the connection environment is uncertain or the candidate strategies are not yet clear; When the value of β approaches infinity, the scoring differences are gradually compressed during normalization, and the final probability distribution tends to concentrate on the strategy with the highest score. The system's strategy selection shows a greedy selection, that is, it is more inclined to strengthen the selection of the currently optimal scoring strategy, which is applicable to scenarios where the connection state is stable or the target strategy is already clear; Therefore, the temperature parameter β essentially controls the entropy level and diversity degree of strategy sampling, achieving a dynamic balance between exploration and exploitation during the strategy selection process.

[0060] To make the temperature adjustment process controllable and stable, the system limits the value range of β between 0.3 and 5.0 and sets a change rate limit mechanism to prevent strategy oscillations caused by too rapid fluctuations of β due to drastic environmental changes or misjudgments.

[0061] Step S4: According to the probability distribution, perform a probability sampling once from multiple candidate connection strategies to determine the target connection strategy for the current period.

[0062] Regarding step S4, to avoid single greedy behavior or path dependence during the strategy selection process, the system adopts a probability sampling mechanism to perform a sampling operation among multiple candidate connection strategies based on this probability distribution to determine the target connection strategy for the current period. Specifically, within the current evaluation period, the system, according to the sampling probabilities P1, P2,..., P n of each strategy output by the Softmax function, performs a probability-driven sampling through methods such as random number generation and cumulative distribution matching to extract a strategy as the target connection strategy for the current period.

[0063] To balance the controllability and stability of strategy selection, the system automatically triggers a strategy sampling action at a set period (such as every 10 minutes) to adapt to the dynamic changes in the network environment and improve the timeliness of sampling. The sampling is not a deterministic selection based on the item with the maximum score value, but a sampling based on the normalized probability value of the strategy, so that strategies with lower scores but not inferior ones have the possibility of being tried, maintaining a certain exploration ability of the system.

[0064] Step S5: Determine whether the target connection strategy is consistent with the currently connected strategy of the cellular communication module. If not, control the cellular communication module to disconnect the current connection and establish a network access path corresponding to the target connection strategy.

[0065] Regarding step S5, specifically, the system first judges the consistency between the target connection policy determined in the current cycle and the actual connected policy where the cellular communication module is currently located, including one-by-one comparison of fields such as whether the base station cell identifier (Cell ID) is the same, whether the current frequency band combination (Band) is consistent, whether the operator identifier accessed has changed, and whether the radio access technology type (RAT, such as LTE, NR, etc.) is consistent.

[0066] If the judgment result is consistent, that is, there is no difference between the target policy and the current policy, the system does not perform any switching operation, the current connection remains unchanged, and it enters the evaluation process of the next cycle.

[0067] If the judgment result is inconsistent, it indicates that there is a deviation between the current connection state and the optimization result. The system will perform connection switching control operations, specifically including: Disconnect the current connection: Control the cellular communication module to safely disconnect the current network access path through the standard disconnection process (such as releasing the PDP context, deregistering the cell, etc.); Execute reselection connection: According to the target cell, frequency band, operator, and access type specified in the target connection policy, reconstruct the network access parameters, and complete connection processes such as network reselection, registration, and activation; Status update and caching: After the switching is successful, update the current connection state information, and record the switching result for policy feedback and model update in subsequent cycles.

[0068] The above judgment and switching mechanism can prevent the cellular communication module from frequently switching networks when the policy score fluctuates slightly, improving the execution stability of the policy sampling behavior and the convergence speed of the connection path selection.

[0069] Step S6: According to the performance evaluation result of the current connection state, dynamically adjust the temperature parameter for calculating the probability distribution in the normalization process, so as to dynamically balance exploration and exploitation in the connection policy selection, thereby optimizing the sampling behavior of the connection policy in the next cycle.

[0070] Regarding step S6, specifically, the system dynamically adjusts the temperature parameter used to generate the candidate policy sampling probability distribution in the normalization process based on the connection state performance of the cellular communication module in the current cycle. This temperature parameter, as a regulatory factor in the score normalization function, its change will directly affect the sampling probability dispersion degree of each candidate connection policy, and thus affect the tendency and diversity of policy selection. Through the adaptive regulation of the temperature parameter, the system can dynamically balance exploration (i.e., expanding the candidate policy range) and exploitation (i.e., concentrating on selecting policies with better scores) in policy selection according to the connection stability, signal quality, or policy execution effect in different network environments. This adjustment mechanism can reduce unnecessary switching while maintaining the flexibility of the connection policy response, and improve the actual effect of policy decision-making.

[0071] In addition, the adjustment process of the temperature parameter is based on the key performance indicators during the operation of the cellular communication module, has dynamic response capabilities and environmental adaptability, can continuously track the change trend of the connection quality and make adjustments, so as to effectively improve the judgment accuracy of policy switching and the overall network access stability of the module in a changing cellular network environment.

[0072] It is not difficult to find that, compared with the related technologies, in the solution provided by the embodiment of the present application, a weighted scoring model is introduced to quantitatively evaluate multiple candidate connection policies and perform normalization processing to form a probability distribution; a temperature parameter is introduced as a control factor to adjust the discreteness of the sampling probabilities between candidate connection policies during the scoring normalization process, realizing dynamic balance control between exploration and exploitation during the connection policy selection process. When the connection state is good, the system strengthens the sampling tendency of high-scoring policies to improve connection stability; when the environment fluctuates or the signal deteriorates, the policy selection range is expanded to improve robustness and recovery ability, thus significantly improving the connection stability, policy switching flexibility and overall network access quality of the cellular communication module.

[0073] Second Embodiment The second embodiment of the present application relates to a network connection optimization method for a cellular communication module. The second embodiment is an improvement based on the first embodiment. The specific improvement lies in: in the second embodiment of the present application, a specific implementation manner for calculating the weighted score of candidate connection policies is provided, that is, step S2 may further include the following steps: Step S201: For each candidate connection policy, according to its multi-dimensional feature vector related to the current network state, a set of weight values is respectively configured. A set of weight values corresponds one by one to each standardized value in the multi-dimensional feature vector, and is used to represent the response sensitivity or priority consideration degree of the candidate connection policy to each parameter information. The value range of the weight value is between 0.01 and 1, which is used to prevent the weight from failing or being over-amplified; Step S202: Multiply each standardized value in the multi-dimensional feature vector by the corresponding weight value, and sum up all the product results to obtain the score value of the candidate connection policy under the current network state.

[0074] For each candidate connection policy constructed in the current cycle, the system respectively configures a set of corresponding weight value sets according to its corresponding multi-dimensional feature vector. The multi-dimensional feature vector includes multiple feature components obtained by standardizing network state parameters, and each set of weight values is used to represent the response sensitivity or priority tendency of the candidate connection policy to each parameter dimension in the current network environment.

[0075] Exemplarily, each candidate policy corresponds to a set of weight values , whose elements correspond one-to-one with the eigenvector , where n represents the number of feature dimensions. Since in the score calculation, the score of each candidate connection strategy is obtained by summing the products of multiple parameter feature values and the corresponding weight values. If the weight of a certain feature is set to 0, then no matter what the value of this feature is, the product will be 0, and this feature will not participate in the score calculation at all, that is, the weight becomes zero and fails. On the contrary, if the weight of a certain feature is set very large, such as much larger than other features, then after normalization or Softmax processing, this feature will dominate the final score too much, that is, the weight is extremely high, resulting in score imbalance and the strategy selection relying too much on a single feature, affecting the generalization ability and stability of the system. Therefore, the value range of each weight value is set between 0.01 and 1 to prevent the feature weight from becoming zero and failing or having an extremely high weight after normalization.

[0076] The above weight values can be preset through static configuration. When the system is deployed, a set of fixed weight values is set, for example: set the signal strength to 0.8, indicating great attention; set the frequency band type to 0.4, indicating average importance; set the connection stability to 0.6, indicating above average.

[0077] It is also possible to adopt a reinforcement learning mechanism, such as using a deep Q-network, to automatically learn the importance weights of each feature under each strategy. Specifically, the system continuously tries different strategies (connecting different network combinations) during the actual operation process; after each attempt, observe the connection effect, such as whether the connection is successful, whether it drops the line, and the speed; then reward and punish the weight configuration according to the results of these attempts. After repeated attempts for a period of time, the system can learn which features are more important and give the corresponding weights. In this way, the attention degree of each strategy to different feature dimensions can be gradually optimized during the long-term operation process.

[0078] The system calculates the score value of the candidate strategy. The calculation method is to multiply each component in the eigenvector by its corresponding weight value and sum all the product results. That is:

[0079] Among them, is the score value of the candidate connection strategy i in the current network state; is the k-th component (normalized) of the current eigenvector; is the weight value of strategy i for feature .

[0080] For the current strategy, if the network feature performance (i.e., ) in a certain dimension is good, and the importance weight of this dimension (i.e., If it is higher, this dimension will have a more positive impact on the final score. On the contrary, if a certain feature performs poorly but has a low weight, its impact on the score will be weakened, thus ensuring the balance, adjustability, and interpretability of the score.

[0081] Combine the scoring structures of all candidate strategies into a scoring matrix , which is used for subsequent normalization processing and probability sampling operations. The range of the score values will be within a limited range due to the normalization interval of the input features (usually 0 to 1) and the weight constraints (such as 0.01 to 1.0), ensuring good stability during subsequent normalization processing.

[0082] It is not difficult to find that in the embodiments of the present application, by introducing a scoring calculation mechanism based on the weighted sum of the feature vector and the weight vector, this mechanism allows a set of weight values corresponding one-to-one to the multi-dimensional feature vectors to be configured for different candidate connection strategies respectively, so as to realize the differential response of the strategies to different network parameter dimensions. This fine-grained weight allocation method can more accurately reflect the performance differences of each strategy in different network environments. Especially in scenarios with multiple access paths, large parameter fluctuations, or frequent dynamic changes in the environment, it can improve the resolution and stability of the score, and avoid excessive averaging or bias of the strategy scores.

[0083] The third embodiment The third embodiment of the present application relates to a network connection optimization method for a cellular communication module. The third embodiment is an improvement based on the first embodiment. The specific improvement lies in: in the third embodiment of the present application, a specific implementation manner of dynamically adjusting the temperature parameter based on the connection state evaluation result is provided, that is, step S6 can further include the following steps: Step S601: In each preset evaluation period, count the connection state indicators of the cellular communication module. The connection state indicators include the historical disconnection times, the change rate of the comprehensive signal quality score, and the success rate of connection strategy switching; Step S602: Calculate the disconnection rate indicator according to the historical disconnection times. The disconnection rate indicator is the normalized average value of the historical disconnection times within the evaluation period, and is used as a control factor to participate in the calculation of the temperature parameter; Step S603: Set the temperature parameter as the sum of the base value and the compensation term, where the base value is a fixed reference value, and the compensation term increases as the disconnection rate indicator decreases; Step S604: When it is detected that the change rate of the comprehensive signal quality score in the evaluation period drops by more than the preset percentage threshold compared with the previous evaluation period, reduce the temperature parameter by the preset first adjustment factor, so that the sampling probability distribution after score normalization tends to be balanced, and increase the selection probability of non-optimal connection strategies to enhance the exploration of connection strategies; Step S605: When the success rate of connection strategy switching is detected to exceed the preset success rate threshold within the evaluation period, the temperature parameter is amplified according to the preset second adjustment factor, making the sampled probability distribution after score normalization more concentrated, and preferentially sampling candidate connection strategies with high scores to enhance the exploitativeness of the connection strategy.

[0084] Specifically, within each preset policy evaluation period (for example, every 10 minutes), the system real-time statistics multiple connection status indicators of the cellular communication module, including the historical disconnection times, the change rate of the comprehensive signal quality score, and the success rate of connection strategy switching, etc., to reflect the connection stability and policy response effect of the module in the current environment.

[0085] Based on the disconnection times recorded within the period, a normalized disconnection rate index is calculated as an important reflection of connection stability, and this index is used as a control factor to participate in the calculation process of the temperature parameter β. Specifically, the temperature parameter β is defined as the sum of a base value and a compensation term. The base value is a fixed reference value (such as 1.0), and the compensation term increases as the disconnection rate index decreases, that is, the fewer disconnections and the more stable the network, the higher the temperature parameter, thereby strengthening the sampling tendency of high-quality strategies.

[0086] When it is detected that the decrease amplitude of the comprehensive signal quality score in the current period compared with the previous period exceeds the preset percentage threshold (such as 20%), the system determines that the current network connection quality may have a deteriorating trend, or the current policy model has insufficient adaptability to the environment. To improve the diversity and recovery ability of policy sampling, the system reduces the temperature parameter β according to the preset adjustment factor (for example, multiplied by 0.8), so that the policy probability distribution after score normalization becomes smoother, increasing the sampled probability of sub-optimal strategies or unselected strategies, and enhancing the exploratory nature of policy selection.

[0087] When it is detected that the success rate of policy switching in the current period exceeds the set threshold (such as 95%), it indicates that the judgment result of the current score model on the quality of the strategy is highly consistent with the actual execution effect, and the policy sampling has a high hit rate. At this time, the system increases the temperature parameter β (for example, multiplied by 1.2), making the normalization result further focus on the strategies with high scores, preferentially sampling candidate connection paths with better scores, thereby enhancing the exploitativeness of policy selection and accelerating the convergence and performance stability of the connection path.

[0088] It is not difficult to find that in the embodiments of the present application, by periodically collecting key performance indicators such as the disconnection times, the change trend of signal quality, and the success rate of policy switching during the actual operation of the cellular communication module, the effectiveness of the current connection policy and the stability of the network environment can be accurately perceived, and accordingly, the value range and change direction of the temperature parameter in the score normalization process are dynamically adjusted, thereby affecting the sampling probability distribution of candidate policies. This makes the policy selection process have the characteristics of both flexibility and goal orientation, and significantly improves the connection optimization efficiency and robustness of the cellular communication module in a dynamic network environment.

[0089] It should be noted that the third embodiment of the present application can also be an improvement based on any one or more of the first embodiment to the second embodiment.

[0090] Fourth Embodiment The fourth embodiment of the present application relates to a network connection optimization method for a cellular communication module. The fourth embodiment is an improvement based on the first embodiment. The specific improvement lies in: in the fourth embodiment of the present application, a specific implementation manner of adaptively adjusting the weight value in the scoring model by combining a continuous disconnection event judgment mechanism is provided, that is, the network connection optimization method further includes: When it is monitored that the disconnection event is a continuous disconnection event, in addition to dynamically adjusting the temperature parameter, the weight value used in the score calculation process is also adaptively adjusted. The determination conditions for the continuous disconnection event include any one of the following conditions being satisfied: The number of disconnections within a set time period exceeds a preset number threshold, or, within a set time window, the number of failures to receive the base station heartbeat message by the cellular communication module exceeds a preset failure number threshold, or, the received signal power continuously remains lower than a preset threshold and the duration exceeds a set duration. In the case where it is determined as a continuous disconnection event, the system performs an adaptive adjustment operation on the weight value, specifically including: increasing the weight value related to the operator network service quality index and / or decreasing the weight value related to the working frequency band type; In the case where it is not determined as a continuous disconnection event, the system enters a conventional incremental learning process to maintain the continuous learning ability of the scoring model of the candidate connection policy.

[0091] Specifically, referring to Figure 3As shown, if a certain feature becomes extremely unstable or fails within a short period of time. For example, there is long-term interference in a certain frequency band, a local anomaly of a certain operator in this area, the cell handover frequency loses its discrimination ability, etc. At this time, if the scoring model still maintains the original high weight, it will overly rely on a judgment basis that has actually failed, resulting in incorrect policy selection. Therefore, when the system detects that a disconnection event constitutes a continuous disconnection event, in addition to dynamically adjusting the temperature parameter β based on the connection performance indicators, it further makes targeted adjustments to the feature weight values used in the scoring model to correct the tendency of overly relying on unstable parameters during the scoring process.

[0092] The determination conditions for the continuous disconnection event include any of the following situations: Within a set time period (such as greater than 60 seconds and less than 20 minutes), the number of disconnections exceeds a preset number threshold (such as more than 3 times); Within a set time window, the number of failures of the cellular communication module to receive the base station heartbeat message exceeds a preset failure number threshold; The received signal power RSRP of the module continuously remains below a preset threshold (such as -110dBm), and this state lasts for a set duration (such as 30 seconds).

[0093] Once it is determined as a continuous disconnection event, the system will execute the adaptive adjustment strategy of the weight value in the scoring model, specifically including: increasing the weight value related to the operator network service quality indicator to enhance the sensitivity of the scoring model to the stability performance of the operator; decreasing the weight value related to the current working frequency band type to weaken the dependence on the frequency band priority configuration and avoid policy selection imbalance in high-interference or edge areas. In this way, the adaptive correction of the scoring mechanism is realized, making it more suitable for the current network environment, thereby improving the accuracy of the next-round connection policy evaluation and selection.

[0094] If a continuous disconnection event is not triggered during the current evaluation period, that is, the connection state of the cellular communication module is overall within the stable range, the system will enter the regular incremental learning process. Specifically, the system makes fine-tuning on the feature weights, sampling feedback, or historical values of temperature parameters in the scoring model based on the connection performance data (including policy scores, actual connection results, success rates, etc.) collected in the current period. This update adopts a progressive learning strategy, that is, only a limited proportion of new data is introduced each time to correct the existing model parameters, thus effectively avoiding model oscillation or unstable policy selection caused by short-term fluctuations. In this way, without interrupting the normal operation of the policy scoring mechanism, lightweight parameter updates are made to the scoring model using the evaluation data accumulated in the current period to achieve the gradual optimization of the model and the continuous improvement of the environment adaptation ability.

[0095] It is not difficult to find that in the embodiment of the present application, compared with the method of sampling based only on the scoring value and temperature parameter control strategy, the response path to abnormal network behavior is further refined: when the system continuously detects typical connection failure modes such as disconnection, heartbeat loss or long-term low signal, it can accurately identify and actively adjust the emphasis on different parameter dimensions in the scoring logic, fundamentally guiding the scoring results to be closer to the actual connection performance. The self-healing ability of the cellular communication module in the case of abnormal network connection and the pertinence of the connection strategy adjustment are further enhanced.

[0096] It should be noted that the fourth embodiment of the present application may also be an improvement based on any one or more of the first to third embodiments.

[0097] Fifth embodiment The fifth embodiment of the present application relates to a network connection optimization method for a cellular communication module. The fifth embodiment is an improvement based on the first embodiment, and the specific improvement is: in the fifth embodiment of the present application, a specific implementation method based on the sampling range of the candidate priority queue restriction strategy is provided, that is, before step S4, the network connection optimization method also includes: According to the normalized result of the score value, the candidate connection strategies with the highest score ranking are constructed into a candidate priority queue, and the probability sampling operation is limited to be performed only in the candidate priority queue, where the candidate connection strategies included in the candidate priority queue are the candidate connection strategies with the top k score rankings, where k is an adjustable parameter.

[0098] Specifically, after performing normalization processing such as the Softmax function, the system obtains the scoring probability value corresponding to each candidate connection strategy. Sort all candidate strategies in descending order, and select the top k strategies with the highest scores to form a candidate priority queue. , where k is the set priority sampling window size, which can usually be set to 3~10 and can be flexibly adjusted depending on the total number of candidate strategies and the complexity of the network environment.

[0099] At this time, the subsequent probability sampling step will no longer operate based on the complete scoring probability distribution of all candidate strategies. Instead, it re-normalizes the strategy score values in the priority queue to form a restricted sampling distribution, ensuring that sampling only occurs within the range of high-scoring strategies. That is, it retains the original score proportion of the strategies in the priority queue; it forcibly sets the sampling probability of low-scoring (not within the candidate priority queue) strategies to zero; it recalculates the normalized distribution within the priority sampling range to meet the requirement that the sum of Softmax probabilities is 1. This method can prevent mis-sampling caused by probability tail noise, especially when there is still a long-tail effect of the Softmax function (even if the score differences are significant, the low-scoring items at the tail may still have non-zero sampling probabilities), and it has significant advantages.

[0100] In addition, to improve adaptability, the system can introduce the following adjustable mechanisms, including: k-value dynamic adjustment strategy: When the scoring distribution of candidate strategies is too balanced or fluctuates greatly, the system can dynamically increase k to retain more alternative paths; when the differences between high-scoring strategies are significant, k can be reduced to improve the convergence efficiency.

[0101] Candidate priority queue refresh period: Combining with the actual strategy execution period (such as once every 10 minutes), the priority queue is refreshed periodically to ensure coverage of the latest scoring trends.

[0102] Forced retention mechanism: At least one historically well-performing strategy (such as the strategy with successful connection in the previous period) is retained as a redundancy during queue construction to enhance robustness.

[0103] It is not difficult to find that in the embodiments of this application, by introducing the candidate priority queue mechanism, the system can, on the basis of maintaining strategy diversity, effectively filter out invalid strategies with significantly low scores, reduce the probability of mis-sampling, improve the decision-making reliability of strategy switching, shorten the sampling convergence time, and enhance the goal orientation and effectiveness of connection path selection.

[0104] It should be noted that the fifth embodiment of this application can also be an improvement based on any one or more of the first to fourth embodiments.

[0105] The step division of the above various methods is only for clear description. When implemented, they can be combined into one step or some steps can be split into multiple steps. As long as the same logical relationship is included, they are all within the protection scope of this application; adding insignificant modifications to the algorithm or process or introducing insignificant designs, but not changing the core design of its algorithm and process, are all within the protection scope of this application.

[0106] In addition, some embodiments of the present application further provide an electronic device. The electronic device may be various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and so on. The electronic device may also be various forms of mobile devices, such as personal digital processors, cellular phones, smart phones, wearable devices, and other similar computing devices.

[0107] The electronic device includes: one or more processors; and a memory storing computer program instructions that, when executed, cause the processors to execute a network connection optimization method for a cellular communication module provided in any one or more of the above embodiments. Figure 4 An exemplary structural diagram of the electronic device is disclosed. The electronic device includes: one or more processors 1101, a memory 1102, and interfaces for connecting various components, including a high-speed interface and a low-speed interface. Each component is interconnected using different buses and may be mounted on a common motherboard or otherwise as needed. The processor may process instructions executed within the electronic device, including instructions stored in the memory or on the memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, if needed, multiple processors and / or multiple buses may be used together with multiple memories and multiple memories. Similarly, multiple electronic devices may be connected, and each device provides part of the necessary operations. Among them, the components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0108] The electronic device may further include: an input device 1103 and an output device 1104. The processor 1101, the memory 1102, the input device 1103, and the output device 1104 may be connected by a bus or other means, Figure 4 taking the connection by bus as an example.

[0109] The input device 1103 may receive input digital or character information, and generate key signal inputs related to the user settings and function controls of the electronic device, such as input devices like touch screens, keypads, mice, trackpads, touchpads, pointing sticks, one or more mouse buttons, trackballs, joysticks, etc. The output device 1104 may include a display device, an auxiliary lighting device (such as an LED), and a haptic feedback device (such as a vibration motor), etc. The display device may include, but is not limited to, liquid crystal displays, light-emitting diode displays, and plasma displays. In some embodiments, the display device may be a touch screen.

[0110] To provide interaction with a user, the electronic device may be a computer. The computer has: a display device for displaying information to the user (e.g., a cathode ray tube or an LCD monitor); and a keyboard and a pointing device (e.g., a mouse), through which the user can provide input to the computer. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback); and input from the user can be received in any form (e.g., voice input or tactile input).

[0111] In the embodiments of the present application, a computer program / instructions is stored on a computer-readable medium. When the computer program / instructions are executed by a processor, a network connection optimization method for a cellular communication module provided by any one or more of the above embodiments is implemented. The computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist separately and not be assembled into the device. The above computer-readable medium carries one or more computer-readable instructions.

[0112] The memory 1102 can be used as a non-transitory computer-readable storage medium, and can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules. By running the non-transitory software programs, instructions, and modules stored in the memory 1102, the processor 1101 executes various functional applications and data processing of the server, so as to implement the program instructions / modules corresponding to the method provided by any one or more of the above embodiments in the embodiments of the present application.

[0113] The memory 1102 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the electronic device. In addition, the memory 1102 may include a high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some embodiments, the memory 1102 may optionally include a memory remotely set relative to the processor 1101, and these remote memories may be connected to the electronic device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0114] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, the computer-readable medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, apparatus, or device.

[0115] The computer-readable medium includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of the computer's storage medium include, but are not limited to, phase change memory, static random access memory, dynamic random access memory, other types of random access memory, read-only memory, electrically erasable programmable read-only memory, flash memory, or other memory technologies, compact disc read-only memory, digital versatile disc, or other optical storage, magnetic cassette tapes, magnetic disk storage, or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0116] The computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as C language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network or a wide area network, or can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0117] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. For example, dedicated integrated circuits, general-purpose computers, or any other similar hardware devices can be used. In some embodiments, the software program of the present application can be executed by a processor to implement the above steps or functions. Similarly, the software program of the present application (including related data structures) can be stored in a computer-readable recording medium, such as a RAM memory, a magnetic or optical drive, or a floppy disk and the like. Additionally, some steps or functions of the present application can be implemented using hardware, for example, as a circuit that cooperates with a processor to execute each step or function.

[0118] The computer program product provided by the embodiments of the present application includes one or more computer programs / instructions. When the computer program / instructions are executed by a processor, they wholly or partly generate the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server, data center, etc. that contains one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state drive), etc.

[0119] The flowcharts or block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of devices, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks can occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0120] The scope of the present application is defined by the appended claims rather than the above description. Therefore, all changes that fall within the meaning and scope of the equivalent elements of the claims are intended to be encompassed by the present application. Any reference numerals in the claims should not be construed as limiting the claims involved. In addition, it is obvious that the term "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of elements or devices stated in the apparatus claims may also be implemented by one element or device through software or hardware. The terms "first", "second", etc. are only used for descriptive distinction and do not represent any specific order, nor can they be construed as indicating or implying relative importance.

[0121] As described above, the above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily make changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.

Claims

1. A network connection optimization method for a cellular communication module, characterized in that, Including: Collecting in real time parameter information for characterizing the current network connection status, and performing feature processing on the parameter information to construct a multi-dimensional feature vector, where the parameter information includes: base station cell handover frequency, operating frequency band type, received signal power, signal-to-noise ratio, operator network service quality index, connection stability index; For each of the constructed multiple candidate connection strategies, calculating the score value of each candidate connection strategy based on the multi-dimensional feature vector and a preset weight value respectively; Performing normalization processing on each of the score values to obtain the probability distribution of the candidate connection strategies; According to the probability distribution, performing probability sampling once periodically from the multiple candidate connection strategies to determine the target connection strategy for the current period; Judging whether the target connection strategy is consistent with the currently connected strategy of the cellular communication module. If not, controlling the cellular communication module to disconnect the current connection and establish a network access path corresponding to the target connection strategy; According to the performance evaluation result of the current connection status, dynamically adjusting the temperature parameter for calculating the probability distribution in the normalization processing, so as to dynamically balance exploration and exploitation in the connection strategy selection, thereby optimizing the sampling behavior of the connection strategy for the next period.

2. The network connection optimization method for a cellular communication module according to claim 1, wherein The step of performing feature processing on the parameter information to construct a multi-dimensional feature vector includes: Performing feature engineering processing on the original collected values of the parameter information to convert them into standardized numerical values that can be used for subsequent score calculation, and combining the standardized numerical values into the form of a multi-dimensional feature vector. The feature dimension of the multi-dimensional feature vector can be expanded or trimmed according to the service scenario. Among them, the value range of the standardized numerical value is from 0 to 1, indicating the performance of the candidate connection strategy in the dimension of parameter information.

3. The network connection optimization method for a cellular communication module according to claim 1, wherein Constructing multiple candidate connection strategies specifically includes: Obtaining network access parameters detectable by the cellular communication module in the current network environment, where the network access parameters include: cell identifier, frequency band type, operator identifier, and radio access technology type; Based on different combination methods of the network access parameters, constructing multiple candidate connection strategies, and the candidate connection strategies are used to represent the connection paths that the cellular communication module can select in the current network environment.

4. The network connection optimization method according to claim 2, wherein The step of calculating the score value of each candidate connection strategy based on the multi-dimensional feature vector and a preset weight value respectively for each of the constructed multiple candidate connection strategies includes: For each candidate connection strategy, configuring a set of weight values according to its multi-dimensional feature vector related to the current network state. The set of weight values corresponds one by one to each standardized numerical value in the multi-dimensional feature vector, and is used to characterize the response sensitivity or priority consideration degree of the candidate connection strategy to each parameter information. The value range of the weight value is between 0.01 and 1, which is used to prevent the weight from failing or being over-amplified; Multiplying each standardized numerical value in the multi-dimensional feature vector by the corresponding weight value, and summing up all the product results to obtain the score value of the candidate connection strategy in the current network state.

5. The network connection optimization method for a cellular communication module according to claim 1, wherein The step of dynamically adjusting the temperature parameter of the probability distribution calculated in the normalization process according to the performance evaluation result of the current connection state to dynamically balance the exploration and utilization in the connection strategy selection includes: In each preset evaluation period, the connection status indicators of the cellular communication module are counted, and the connection status indicators include the number of historical disconnections, the change rate of the comprehensive signal quality score, and the connection strategy switching success rate; Calculate the disconnection rate index according to the historical disconnection times, where the disconnection rate index is the normalized average value of the historical disconnection times within the evaluation period, and participates in the calculation of the temperature parameter as a control factor; The temperature parameter is set as the sum of a basic value and a compensation item, wherein the basic value is a fixed reference value, and the compensation item increases as the offline rate indicator decreases; When it is detected within the evaluation period that the rate of change of the comprehensive score of the signal quality decreases by more than a preset percentage threshold compared with the previous evaluation period, the temperature parameter is reduced by a preset first adjustment factor, so that the sampling probability distribution after the score normalization tends to be balanced, and the probability of selecting a non-optimal connection strategy is increased, so as to enhance the exploratory nature of the connection strategy; When it is detected within the evaluation period that the connection strategy switching success rate exceeds a preset success rate threshold, the temperature parameter is amplified according to a preset second adjustment factor to make the sampling probability distribution after score normalization more concentrated, and candidate connection strategies with high scores are sampled first to enhance the utilization of connection strategies.

6. The network connection optimization method for a cellular communication module according to claim 1 or 4, characterized in that Also includes: When the offline event is detected as a continuous offline event, in addition to dynamically adjusting the temperature parameter, the weight value used in the scoring calculation process is also adaptively adjusted. The determination condition of the continuous offline event includes any of the following conditions: The number of disconnections within a set time period exceeds a preset number threshold, or the number of failures of the cellular communication module to receive the base station heartbeat message exceeds a preset number threshold within a set time window, or the received signal power is continuously lower than a preset threshold and the duration exceeds a set duration, In the case of determining the continuous disconnection event, the system performs an adaptive adjustment operation of the weight value, specifically including: increasing the weight value related to the operator network service quality indicator and / or reducing the weight value related to the working frequency band type; In the case where the continuous disconnection event is not determined, the system enters a conventional incremental learning process to maintain the continuous learning capability of the scoring model of the candidate connection strategy.

7. The network connection optimization method for a cellular communication module according to claim 1, characterized in that Before the step of periodically performing a probability sampling from a plurality of the candidate connection strategies according to the probability distribution, the method further includes: According to the normalized result of the score value, the candidate connection strategies with the highest score ranking are constructed into a candidate priority queue, and the probability sampling operation is limited to be performed only in the candidate priority queue, wherein the candidate connection strategies included in the candidate priority queue are the candidate connection strategies with the top k score rankings, where k is an adjustable parameter.

8. An electronic device, characterized in that, The electronic device comprises: One or more processors; and a memory storing computer program instructions which, when executed, cause the processors to perform the network connection optimization method for a cellular communication module as recited in any one of claims 1-7.

9. A computer-readable storage medium having a computer program and / or instructions stored thereon, characterized in that, When the computer program and / or instructions are executed by a processor, the network connection optimization method for a cellular communication module as recited in any one of claims 1-7 is implemented.

10. A computer program product, comprising a computer program and / or instructions, characterized in that, When the computer program and / or instructions are executed by a processor, the network connection optimization method for a cellular communication module as recited in any one of claims 1-7 is implemented.

Images